Per. Mineral. (2004), 73, 113-125 http://go.to/permin SPECIAL ISSUE 3: A showcase of the Italian research in applied petrology

An International Journal of MINERALOGY, CRYSTALLOGRAPHY, GEOCHEMISTRY, ORE DEPOSITS, PETROLOGY, VOLCANOLOGY and applied topics on Environment, Archaeometry and Cultural Heritage

Archaeometric aspects of white and coloured used in antiquity: the state of the art

LORENZO LAZZARINI*

Laboratorio di Analisi dei Materiali Antichi, Dipartimento di Storia dell’Architettura, Università IUAV di Venezia, S. Polo 2468 – 30125 – Venezia.

ABSTRACT. — As is well known, the identification metodologia. Attualmente le più alte probabilità di of the quarry of ancient artefacts is of the successo si raggiungono combinando due o più utmost interest to archaeologists and art historians. tecniche analitiche. Nell’articolo si presenta una Scholars of different disciplines have been trying for breve storia degli studi, e vengono considerate le more than a century such an identification by means tecniche e le combinazioni più usate, in particolare of a unique laboratory analysis without positive quella dell’esame petrografico di sezioni sottili e results. To-date the best probabilities of success are l’analisi degli isotopi stabili del C e dell’O che si obtained by combining together two or more avvantaggia della più completa banca dati di analytical techniques. The paper presents a short riferimento esistente per i marmi più importanti usati history of the important studies, namely those in antico. Di quest’ultimi, oltre ad alcune essenziali regarding the most frequently used combinations, informazioni circa le località di cava, i periodi di uso with particular reference to the petrographic e la bibliografia specifica più recente, si riportano i examination of a thin section and the determination diagrammi dell’MGS (Maximum Grain Size) e of the C & O stable isotopic ratios on the same isotopici più utili per la loro identificazione. sample. Such a combination takes advantage of the best existing database for the marbles most KEY WORDS: ancient white and coloured marbles, commonly used in antiquity. Of these are reported quarries, archaeometric identification, state of the the essential information on their quarries, periods of art. use and specific bibliography, together with the MGS (Maximum Grain Size) and isotopic diagrams useful for their identification. INTRODUCTION

RIASSUNTO. — Come è noto, l’identificazione della cava di manufatti marmorei antichi riveste un The first scholar to take an interest in the grande interesse archeologico e storico-artistico. Da problem of identifying marbles of oltre un secolo si tenta di raggiungerla con indagini Mediterranean origin that were used in ancient di laboratorio le più varie, senza peraltro essere times, for the purposes of art history as well as giunti a risultati univoci impiegando una sola archaeology, was Winkelmann, as far back as the end of the XVIII century. His interest was * E-mail: [email protected]. taken up by whole generations of scholars but 114 L. LAZZARINI it was not until a hundred years later that R. area of production; monuments (latu sensu) that Lepsius (Lepsius, 1890) developed the first are properly dated can in turn tell us when scientifically correct approach, one that can certain quarries were being worked and hence unreservedly be defined as archaeometric in the the demand for the marbles concerned; the strict modern sense of the term. identification of marbles forming part of sunken What follows is a rapid historical overview cargoes enables ancient trade routes to be of archaeometric studies applied to true reconstructed; location of the quarry from which marbles, white, grey and coloured, dwelling a damaged marble came makes it possible to more on the methods currently most frequently find sound material for the purpose of used to solve the «problem of the provenance» restorations, replacements, copies, etc.. of marble artefacts and focusing in particular So far the many studies that have set out to on the methodology followed by the author in determine the provenance of marbles used in the study of many ancient marble monuments antiquity have led to results which are partial and objects, including those at Ostia, examined and not always satisfactory from a scientific in the framework of what is certainly the most point of view, especially when identification extensive archaeometric survey so far to have was based on macroscopic examination, on been undertaken. visual evidence such as the colour, brilliance or Marbles in the proper sense of the word, that grain of the stone. Such identifications almost is pure carbonatic (calcitic and/or dolomitic) always turn out to be wrong (Renfrew and rocks with a carbonate content that is usually Springer Peacy, 1968), essentially because it is well in excess of 95% – they are crystalline, often found that samples of marble from the they may be white or grey or red or green and same quarry have different autoptic properties they will have been produced by contact or while others from different quarries, sometimes regional metamorphism – are quite common a long way from each other, are identical. throughout the Mediterranean area. Many of More reliable are identifications based on the them were used in antiquity, both in pre- and mineralogical and petrographic study of thin early-historical periods (late Neolithic – sections (Herz and Pritchett 1953, Herz 1955, Cycladic civilizations) and also in the Greek Weiss 1954, Renfrew and Springer Peacy and Roman and later times, when the quantity 1968, Young and Ashmole 1968); some of of marble employed and the quality of the these are often difficult and time-consuming to workmanship often reached remarkable levels. perform however (e.g. petrotectonic In the case of some marbles we know when examinations), and they cannot always cope they were used for the first time and we have with the sometimes considerable local information from various sources that enables variability in mineralogical (e.g. with the us to build up at least a partial picture of their contents of dolomite) and petrographic (e.g. distribution and how they were traded and granulometric) characteristics of pure marbles, transported. In most cases however, we really even though pure marbles are among the most know very little, mainly because of the homogeneous mono-mineralic rocks known. fundamental difficulty of identifying marbles Better results have been achieved by reliably when they are found as original geochemical studies of the trace elements structural or decorative elements of ancient present in marble samples taken from ancient buildings or in , or when they are re- quarries. For example, certain Anatolian used in Mediaeval or Renaissance monuments marbles can be distinguished from their Attic (e.g. in Rome and Venice). counterparts on the basis of sodium and It is of great importance to archaeologists that manganese content (Rybach and Nissen 1964), the identification of marbles should be based on but this criterion cannot be used to discriminate scientific data: sculptures of uncertain attribution between Attic marbles and those from the can be ascribed to a specific artist or «atelier» or Cyclades. Archaeometric aspects of white and coloured marbles used in antiquity: the state of the art 115

Another problem is that in the case of both Another early pluridisciplinary contribution Na and Mn there are often considerable involving a combination of petrographic quantitative variations in different samples characteristics (average grain size, type of from the same quarry or in different parts of a crystal shape and structure, semi-quantitative single sample. This can be explained by the assessment of accessory minerals) and the presence of concentrations of these elements in determination of the Calcium/Strontium ratio minerals other than calcite, e.g. epidotes and was proposed by Lazzarini et al. in 1980 micas, the distribution of which in the structure (1980a). This geochemical ratio was of the marble may be quite casual. This makes considered especially important in pure the identification of marble considerably more marbles for various reasons. Strontium is an difficult in cases where just one small sample is isomorphous element of calcium, which it can available. A more recent study (L. Conforto et replace by up to 4% in aragonite (this al. 1975), which examined only marbles of strontium-rich aragonite has been given the Anatolian and Italian origin and determined name mossotite), while it rarely exceeds 1% in their contents in terms of eight trace elements calcite (Chilingar et al. 1967). Most of the (K, Fe, Mn, Si, Ba, Al, Sr and Ti), has given marbles in this study are the product of the partial results, similar to those of the preceding metamorphism of organogenic limestones or in technique, and has thus enabled us to any case contain the remains of fossil characterize certain quarries with respect to organisms, as was also shown by petrographic others. studies of the marbles of Attica (Marinos, But ultimately, the limits of a method of 1948) and the island of Lesbos (Lazzarini et identification that is based on the determination al., 1999). Now, the protective shell or skeleton of oligo-elements lie in the fact that their of many organisms is made of aragonite, the interval of variation is frequently the same for preference for this polymorphous form of samples from different petrographic regions. CaCO3 rather than calcite being regulated by The approach based on measurement of the temperature. Temperature also controls the isotopic ratios of carbon and oxygen has looked Ca/Sr ratio, which remains unaltered in marbles interesting and promising ever since its first (metamorphism being an essentially appearance: it enabled Craig and Craig (1972) isochemical process) so this datum can be to distinguish Greek marbles from Attica from useful in estimating the thermic level reached their counterparts in the Cyclades and L. by these rocks. Soviet scholars (Vinogradov et Manfra et al. (1975) to distinguish between al., 1952) showed that the Ca/Sr ratio is higher marble from various localities in Asia Minor. in carbonatic rocks the closer they are to the Despite the growth of the isotopic data bank Precambrian; a further scientific reason why its (Herz, 1988), however, it became increasingly determination in marbles is of interest. It has clear that neither autoptic observation nor been observed, in fact, that strontium levels petrographic and geochemical studies alone vary considerably in some of the most would enable conclusive identification of the important marbles of antiquity and this has led most famous marbles used in antiquity, while a to the identification of numerous artefacts combination of methods seemed more (Lazzarini et al., 1980b; Lazzarini et al., 1988). promising. The first proposal came from Another important analytical contribution Renfrew and Springer-Peacy, who suggested was made by Cordischi and his co-authors combining a cathode-luminescence study with (1983), with the proposal to use electronic spin a petrographic study of thin sections. In actual resonance (ESR) on the traces of Mn in fact the two authors did not develop these marbles. As the method was applied to an techniques and it was only later, in the 1990s, increasing number of marbles, the initial that the cathodomicrofacies (infra) of several encouraging results were placed in another marbles were defined. perspective by the overlapping of 116 L. LAZZARINI representative fields (Lloyd et al., 1985) – the point of view. In addition to what is revealed same thing happened subsequently with by the main critical study of the literature isotopic analyses and the analyses of the produced up to then (Mariottini, 1998), it is oligoelements – and it was not until a decade clear that the greatest uncertainty was a later that new improvements and developments consequence of the fact that none of the studies were achieved, especially by associating this had taken into simultaneous consideration all investigative method with others (infra). the marble-producing areas of the Another important proposal involving the Mediterranean, only the best-known ones, application of a combination of analytical, ignoring others that may have been petrographic (including the introduction of insignificant for the Greek period but which MGS – maximum grain size of could well have been significant sources for the calcite/dolomite) and geochemical Romans, not to mention the many other methodologies was made by a large team of quarries that have since come to light and Belgian scientists (Moens et al., 1988). added so much to our knowledge of the Attribution of marble artefacts to their quarries extraction and use of crystalline marbles in the of origin was improved but the method was imperial age. still not absolutely reliable. Finally, in 1989, An important stimulus to scientific Barbin and others (1989) re-proposed the use investigation over the last two decades has of cathodoluminescence, with considerably been provided by ASMOSIA (Association for improved equipment and coupled with the the Study of Marbles and Other Stones in petrographic study of the same thin section; Antiquity) and its members. New methods have this led to the determination of accurate been proposed and many new marbles and cathodomicrofacies for the main marbles of quarries, both major and minor, have been antiquity. The cathodoluminescence of these characterized. One of the most significant new stones is linked to the presence of traces of Mn, methods to emerge was the first to be entirely which enhance it, and of Fe, which reduces it. non-destructive (i.e. that did not depend on the Calcite features an orange or blue procurement of a sample), proposed by Careri luminescence while that of dolomite is red. The and his team in 1992; this was based on the use intensity and the distribution of these of portable laser equipment that recorded the luminescences add parameters that are absorbance/reflecting qualities of quarried sometimes so characteristic that they enable marble, a parameter connected essentially to many (not all) kinds of marble, and even the type of structure, granulometry and different quarries yielding the same marble, to composition (presence of carbonaceous/graphite be distinguished from each other: for example, substances) in a given marble. The method the Parian marble lychnites from Stephani has a achieved good discrimination for some marbles blue luminescence while that of the same and more or less nil for others: since it proved marble from Lakkoi is orange. to be of only partial usefulness right from the Some of these analytical techniques were beginning, the perhaps mistaken decision was applied in monographic studies of important taken not to proceed with development of the marbles, including those from Luni (Herz and methodology. Similar results have been Dean, 1986), Thasos (Herz, 1987) and the recorded very recently (2004) by Biricotti and Cyclades (Germann et al., 1988), all of which Severi using an optical non-destructive still provide substantial help to archaeometric methodology. The method is currently being scientists. studied and improved. In conclusion, until the 1990s the many Considerable interest has been aroused by studies designed to identify the marbles used in the determination of the component elements antiquity led only to results that were partial of rare earths (Barbin et al., 1991) carried out and not always satisfactory from a scientific with various methods including NAA (Meloni Archaeometric aspects of white and coloured marbles used in antiquity: the state of the art 117 et al., 1995), and the new ICP-MS = (1992), Pike (1999) and Goette et al. (1999) on Inductively Coupled Plasma-Mass Spectrometry, Pentelic marble. (Green et al., 2002). Though the same Several new quarries have also been unresolved problems again arise for these trace discovered and characterized in recent years elements (repeatability of results, their including those of the Greek islands of Skyros, comparability when obtained with different Fourni (Lazzarini and Cancelliere, 2000) and analytical methods, content variability at small Tinos (Lazzarini and Antonelli, 2003), and and large scale, etc.), we agree with Matthews Microasiatic areas of marble production such (1997) and Green et al. (2002) that the creation as those of Ionia and Caria, including the of a data bank for REE would be immensely region of Ephesus (Koller et al., in press), useful. Miletus and Heraklea on the lake Latmos Electronic paramagnetic resonance (Peschlow-Bindokat and Germann, 1981). spectroscopy (EPR), the equivalent of ESR, There is no question but that identification of was further developed with the creation of the provenance of white marbles used in bigger data banks, but when used alone it was antiquity remains an issue of fundamental still an unreliable way of attributing a marble interest for archaeologists and art historians artefact to its quarry of origin. At first it was and continues to engage scientists of various coupled with certain petrographic disciplines. Despite the efforts of mineralogists, characteristics (Attanasio and Platania, 2002; petrographers, geochemists, statisticians and Polikreti and Maniatis, 2002; Attanasio, 2003) physicists, who, as we have seen, have used an that could be seen with the naked eye, such as extraordinary variety of analytical methods MGS, but these were obviously not the most over the last century and more, the problem can significant or precise; later it was used together still not be considered wholly and satisfactorily with other analytical procedures such as the resolved: even now there is no non-destructive isotopic technique, and subjected to statistical investigative technique that can quickly and processing (Attanasio et al., 2002). Statistical unequivocally establish the provenance of most processing had already been used in the white marbles. assessment of chemical data, especially of trace elements, but the availability of more sophisticated equipment (specific software and THE ANALYTICAL METHODS CURRENTLY more powerful computers) now ensures that the MOST COMMONLY USED, results are more significant, although not AND THE MOST RECURRENT WHITE MARBLES always reliable. On the isotopic front, the study of the As shown above, the approach most likely to isotopes of strontium (Pentia et al., 2002) also produce reliable results concerning the quarries seems to be a promising methodology, though of provenance of a given marble involves the high cost of the analyses makes it studying the sample with at least two somewhat problematic for large-scale independent analytical methodologies and investigations involving samples from quarries jointly processing all the data obtained. The or artefacts. two methods now most widely used are The last dozen or so years have seen the undoubtedly the minero-petrographic analysis appearance of some very important studies of of thin sections and the isotopic analysis of single marbles, including those by Asgari and oxygen and carbon; when used together the Matthews (1995) on Proconnesian marble, by methods enable most marbles to be identified, Bruno et al. (2002) on the marbles of Thasos, if not with absolute certainty then at least with by Bruno et al. (2000) and Herz (2000) on the a fully acceptable degree of reliability. Both marbles of , by Lazzarini et al. (2002) on techniques require an accurate knowledge of Aphrodisian marble, and by Matthews et al. the parameters considered in the marbles of the 118 L. LAZZARINI quarries worked in antiquity; this can only be Pensabene, 1977). We do not yet have enough acquired via the analysis of a large number of data to be able to characterize other important samples taken either from these quarries or classical marbles such as those of Ephesus and from artefacts of known attribution. In other other places in the Meander Valley (present- words, data banks must be available which are day Menderes), including Tiunta, near Denizli, as representative as possible of the Stratonicea, Milasa, etc., in Anatolia, and mineralogical-petrographic characteristics and Mani, Vrestena, Dolianà, etc., in . the 18O/16O and 13C/12C ratios. For the former, Research on these areas is however ongoing observations have been developed on the basis and it may be assumed that the gaps will of the classic treatises of petrography and shortly be filled. metamorphism (Galwey and Jones, 1963; In the first, more important areas mentioned Kretz, 1966; Spry, 1976); the latter have been above, marble was not extracted from just one expressed respectively as the δ18O and δ13C of quarry but from several, which were often the various marbles used in Greek and Roman some kilometres apart. At Paros, Naxos and times, as first proposed by the Craigs. As we Thasos, for example, various extraction sites know, the isotopic composition of oxygen and have been recognized, including one or more carbon, as for other elements with a low atomic quarries yielding marbles with different number, such as H, N, S, Si and Ca, is minero-petrographic and geo-chemical determined via dedicated mass spectrometry characteristics. To date, four ancient sites have following the methodology introduced by been identified at Paros, all situated in an area McCrea (1950), and conventionally expressed of a few square kilometres in the central- in terms of the «δ unit», defined as: northern part of the island. Some of these marbles belong to different geological δcamp = (Rcamp/Rstd - 1) x 1000 formations. The sites include: 1. the underground quarries of the Stefani where Rcamp and Rstd represent the isotopic Valley, near the village of Marathi, which until ratio considered (18O/16O and 13C/12C the VI century B.C. yielded a much-prized respectively for oxygen and carbon) in the variety of marble known as lychnites, sample and in a suitable reference standard. For conventionally referred to as «Paros-1» by the two elements that concern us, the Moens et al. (1988); international standard adopted is known by the 2. small and medium size open cast quarries initials PBB (calcite of the rostrum of a scattered over the area of Lakkoi and especially Belemnitella Americana of the Pee Dee in the Valley of Chorodaki. The marble from formation of North Carolina (Craig, 1957). these quarries, referred to as «Paros-2» by Recently, Gorgoni et al. (2002a) have Moens et al., (1988), is of excellent quality but proposed considerably updated data banks differs from the preceding variety, not only based on hundreds of analyses relative to the because it has different petrographic and minero-petrographic and isotopic characteristics isotopic features but also visually (it often has of the marbles from the main quarries active in grey patches or areas of foliation); Greek and Roman times: Paros, Naxos, Thasos 3. small, mainly open cast quarries to the SE and Mount Pentelic in Greece, which had long of the Valley of Marathi, from which came a been the subject of historical research and marble with similar characteristics to those of applied petrography (Papageogakis, 1967), Paros-2, referred to as «Paros-3» by some Carrara in Italy (Dolci, 1980), Aphrodisias, authors (Germann et al., 1988; Bruno et al., Afyon and Marmara in Anatolia, present-day 2000; Gorgoni et al., 2000); Turkey; the latter two areas were the source of 4. a small quarry recently discovered by marbles, Docimium and Proconnesium, that Karavos near the village of Kostos (Bruno et were widely used in antiquity (Monna e al., 2000), which contains a relatively small Archaeometric aspects of white and coloured marbles used in antiquity: the state of the art 119 quantity of dolomitic marble as well as the (Proconnesian – 1) and the other from those of predominant calcitic variety. The marble from Çamlik (Proconnesian – 2). this quarry could be called «Paros-4». The salient mineralogical and petrographic At Naxos too several ancient and modern characteristics (structure, boundary of the extraction sites have been identified: the most calcite and/or dolomite crystals, significant important ancient sites are those of Apollonas accessory minerals) of the above-mentioned (a vast area in the extreme north of the island marbles are summarized in Table 1 (Gorgoni et characterized by the presence of deposits of al., 2002, mod.), while the maximum grain size coarse grain marble, some of which are very (MGS) appears in Fig. 1. thick) and those of Melanes-Flerià, a hilly area It should be noted that these data have a about 15 km to the south-west of Apollonas general and not absolute validity (anomalous which is famous for the abandoned remains of compositions and characteristics are always many half-finished kouroi. Essentially, the possible) and that accessory minerals that are marbles from the two sites can be distinguished more or less ubiquitous in all marbles – by means of isotopic analysis (Herz, 1985 and minerals such as quartz, common white-mica Lazzarini-Turi, unpublished data). (muscovite/phengite), graphite/carbonaceous There are many sites that yield the Thasian substances, apatite and ferrous oxides marble, some of which were already active in (haematite and limonites) - have not been taken the VI century B.C.. There was intensive into consideration. Similarly, no account has quarrying on the Alikì promontory on the been taken of the presence of dolomiteis, which northern coast of the island, where exploitation very common in almost all marbles (e.g. in continued at least until the VII century A.D.; this Pentelic, Parian from Karavos, Lunense, large-grain marble, which often featured grey Proconnesian, Thasian from Alikì, etc.) in patches and veins, was exported to Italy, Greece various quantity levels, from trace to several and Asia Minor. The quarries at Cape Fanari, per cent (Lazzarini and Mariottini, 1987). about 3 km to the south-east of Limenas, yielded The petrographic data in Table 1 and Fig. 1 a similar marble to that of Alikì and were are used for comparison with the results of the worked at more or less the same time. Dolomitic study of thin sections of ancient marble marble of excellent quality was obtained from artefacts; this comparison in turn must be tested quarries a few kilometres to the south (at Cape against the direct comparison of sections of Vathy, Saliara); this was much used in the reference quarry marbles with sections from Archaic period and later in Roman times for the artefacts. This latter phase is of the utmost statuary and sarcophaghi (Hermann e Newman, importance in cases of doubtful identification. 1995; Marc, 1995; Wurch-Kozelj and Kozelj, As regards isotopic data, the increasing 1995). The marbles from these sites are referred number of isotopic analyses of ancient marbles to respectively as Thasos-2, Thasos-1 and appearing in the literature in recent years Thasos-3 by Moens et al. (1988). Other small testifies to the huge popularity of the technique quarries such as those located near the Acropolis in archaeometric research. Paradoxically, the supplied marble for local use (Herz, 1987). large quantity of data available has resulted in Several vast extraction sites are known on an extremely complex overall picture: in a the island of Marmara; the marbles they global diagram with coordinates of δ13C vs. yielded had different characteristics but they δ18O, which includes all the data available in were generally of medium grain, with grey the literature relating both to samples collected patches, veins and straight-parallel foliations in the various quarries and to those from from organic/graphite substances. Gorgoni et ancient artefacts of certain provenance, the al. (1998) distinguish two main varieties, fields of variability of the two parameters in the chiefly on the basis of isotopic parameters: one extraction areas under consideration overlap comes from the quarries of Saraylar and Kavala considerably and make the use of this 120 L. LAZZARINI

TABLE 1 The minero-petrographic characteristics essential for the microscopic recognition of the important ancient marbles in thin section.

MARBLE Locality CRYSTAL TYPE FABRIC: most frequent features BOUNDARIES

DOKYMAEN embayed / sutured HE/HO mosaic, with strained crystals APHRODISIAN curved / embayed HE mosaic, sometimes lineated and stressed PROCONNESIAN sutured / embayed HE mortar, often with deformed polysinthetic twins THASIAN ALIKI’ curved to sutured HE mosaic, often lineated and stressed VATHY curved to sutured HO/HE mosaic, sometimes with strained crystals NAXIAN APOLLONA Embayed HE mosaic, with strongly strained crystals MELANES Embayed HE mosaic, sometimes lineated, with strained crystals PARIAN LEFKES curved / embayed HE mosaic, sometimes with fine-grained areas and def. twins KARAVOS Embayed HE/HO lineated, with coarse stressed crystals STEPHANI Curved HE/HO mosaic PENTELIC curved / embayed HO/HE lineated, sometimes mosaic LUNENSE straight / curved HO polygonal with triple points, often mosaic

Fig. 1 – MGS of the Main Mediterranean marbles used in antiquity. Archaeometric aspects of white and coloured marbles used in antiquity: the state of the art 121 investigative method problematic. The problem which feature archaeometric problems of has been at least partly overcome by taking provenance entirely similar to those of white account of an important parameter that can be marbles. The example of rosso antico is obtained from thin section analysis of the instructive: it is a haematitic-marble that has sample, i.e. by MGS. On the basis of the been used ever since the Minoan period and average values of this parameter (MGS greater especially in Roman imperial times, when it or lesser by 2 mm, which more or less was given the name marmor taenarium separates the fine-grain marbles from those because of its origins in the vicinity of Cape with a medium-coarse grain), the isotopic Tainaron on the Mani peninsula in the characteristics of the marbles from the various Peloponnese (Greece). Both the evenly sites can be represented by the diagrams of coloured variety and the white veined sort due Figs. 2 and 3 (Gorgoni et al., 2002a). to its wavy scistosity, can easily be mistaken for the analogous types of cipollino rosso, DETERMINATION OF THE PROVENANCE called marmor Iassense by the Romans OF RED MARBLES USED IN ANTUQITY because it was quarried near the the town of Iasos (now Kiykislacik, province of Milas, The scenario described above also applies, Turkey) in the ancient region of Caria. Both with the same conclusions, at least as far as is these marbles were used for architectural currently known, to some coloured marbles, elements and statuary and it is of considerable

Fig. 2 – Global isotopic diagram (including quarry and artefact data) relating to ancient marbles with MGS < 2mm. C = Lunense; D = Docimaenian; Hy = Hymettian; Pa-1 = Parian lychnites; Pa-3 = Parian from Karavos; Pe-1 e 2 = Pentelic. 122 L. LAZZARINI

Fig. 3 – Isotopic diagram as in Fig.2, but including marbles with MGS > 2mm. Aph = Aphrodisian; N = Naxian; Pa-2 = Parian from Lefkes; Pr-1 = Proconnesian from Saraylar, Pr-2 = Proconnesian from Çamlik; T-1 = Thasian from Phanari, T- 2 = Thasian from Alikì, T-3 = Dolomitic Thasian from Vathy-Saliara.

Fig. 4 – General Fe-Ni correlation for the two important red marbles used in antiquity (from Mani and Iasos), to which has been added the much more rare red from Aphrodisias. Archaeometric aspects of white and coloured marbles used in antiquity: the state of the art 123 importance that they should be distinguishable of monuments in the Mediterranean basin», D. from each other, especially for Late Antiquity Decrouez, J. Chamay, F. Zezza (eds), Genève, 47- and the Byzantine period. A first distinction 62. based on isotopic analysis and trace elements, BRUNO M., LAZZARINI L., SOLIGO M., TURI B. and VARTI-MATARANGAS M. (2000) — The ancient which had seemed possible (Gorgoni et al., quarry at Karavos (Paros) and the 1992), was subsequently shown to be characterization of its marble. In «Paria Lithos: inadequate (Lazzarini, in press), thus Parian quarries, marble and workshops of prompting the need for supplementary », D.U. Schilardi, D. Katsonopoulou chemical and statistical analysis. This (eds.), Athens, 95-103. supplement takes into consideration a higher CARERI G., LAZZARINI L. and MAZZACURATI V. (1992) — Angular distribution of light diffused number of reference samples, themselves from laser-irradiated crystalline marbles. analysed for trace elements, taken from the two Potential use for identification purposes. In ancient quarries. The results of all the «Ancient stones: quarrying, trade and analytical series shows that the Fe/Ni provenance», M. Waelkens, N. Herz, L. Moens relationship (Fig. 4) is of itself sufficient to (eds), Leuven, 237-242. solve the archaeometric problem (Gorgoni et CHILINGAR G.V., BISSEL H.J. and FAIRBRIDGE R.W. (1967) — Carbonate rocks-origin, occurrence al., 2002b). and classification, Amsterdam. CONFORTO L., FELICI M., MONNA D., SERVA L. and TADDEUCCI A. (1975) — A preliminary evaluation REFERENCES of chemical data (trace elements) from classical marble quarries in the Mediterranean, ASGARI N., MATTHEWS K.J. (1995) — The stable Archaeometry 17, 201-213. isotope analysis of marble from Proconnesus. In CORDISCHI D., MONNA D. and SEGRE A.L. (1983) — «The study of marble and other stones used in ESR analysis of marble samples from antiquity», Y.Maniatis, N.Herz, Y. Basiakos Mediterranean quarries of archaeological (eds), London, 123-129. interest. Archaeometry, 25, 68-76. ATTANASIO D. (2003) — Ancient white marbles, CRAIG H. (1957) — Isotopic standards for carbon Rome. and oxygen and correction factor for mass- ATTANASIO D., CONTI L., PLATANIA R. and TURI B. spectrometric analysis of carbon dioxide. (2002) – Multimethod provenance Geochim. Cosmochim. Acta, 12, 133-149. determinations: isotopic, ESR and petrographic CRAIG H. and CRAIG V. (1972) — Greek marble discrimination of fine-grained white marbles. In determination of provenance by isotopic analysis, «ASMOSIA VI, Interdisciplinary Studies in Science, 176, 401-403. Ancient Stone», L. Lazzarini (ed.), Padova, 141- DOLCI E. (1980) — Carrara, cave antiche. Carrara. 147. GALWEY A.K. and JONES K.A. (1963) — Attempt to ATTANASIO D., PLATANIA R. and ROCCHI P. (2003) determine the mechanism of a natural mineral- — The most important white m,arbles in Roman forming reaction, Chem. Soc. 1084. architecture: ESR identification and bootstrap GERMANN K., GRUBEN G., KNOLL H., VALIS V. and assessment of the results. In Proc. of the Conf: WINKLER F.J. (1988) — Provenance «Marble in the Roman world», Boulder, April characteristics of Cycladic (Paros and Naxos) 2001, in press. marbles. A multivariate geological approach. In BARBIN V., RAMSEYER K., DECROUEZ D. and HERB «Classical marble: geochemistry, technology, R. (1989) — Marbres blancs: caractérisation par trade», N. Herz, M. Waelkens (eds.), Dordrecht, cathodoluminescence. C.R.Acad Sci Paris, 308, 251-262 Série II, 865-866. See also BARBIN V., RAMSEYER GOETTE H.R., POLIKRETI K., VAKOULIS T. and K., DECROUEZ D., BURNS S.V. and MAIER J.L. MANIATIS Y. (1999) — Investigation of the blue- (1992) — Cathodoluminescence of white marble: gray marble of Pentelikon and the equivalent an overview. Archaeometry, 34, 175-183. Hymettian: possible used in antiquity. In BARBIN V., BURNS S.J., DECROUEZ D., ODDONE M. «Archeomateriaux, marbres et autres roches», and ZEZZA U. (1991) — Cathodoluminescence, M.Schvoerer (ed.), Bordeaux, 83-90. REE, and stable isotopes characterization of GORGONI C., KOKKINAKIS I., LAZZARINI L. and marbles from Crevola D’Ossola, Candoglia, Lasa MARIOTTINI M. (1992) — Geochemical and (Italy) and Pteleos (Greece). In «The conservation petrographic characterization of «rosso antico» 124 L. LAZZARINI

and other white-grey marbles of Mani (Greece). calcite/dolomite. In «Materiali Lapidei», Suppl. al In ASMOSIA 2, «Ancient Stones: quarrying, N°41 del Bollettino d’Arte, 69-72. trade and provenance», Acta Archaeologica LAZZARINI L., MOSCHINI G. and STIEVANO B.M. Lovaniensia, Mon. 2, M. Waelkens, N. Herz, L. (1980a) — A contribution to the identification of Moens (eds.), Leuven, 155-165. Italian, Greek and Anatolian marbles through a GORGONI C., LAZZARINI L., PALLANTE P. and TURI B. petrological study and the evaluation of Ca/Sr (2002a) — An updated and detailed ratio, Archaeometry 22, 173-182. mineropetrographic and C-O stable isotopic LAZZARINI L., MOSCHINI G. and STIEVANO B.M. reference database for the main Mediterranean (1980b) — Alcuni esempi di identificazione di marbles used in antiquity. In «ASMOSIA 5: marmi antichi mediante uno studio petrografico e Interdisciplinary Studies on Ancient Stone», J.J. la determinazione del rapporto Ca/Sr. Quaderni Herrmann, N. Herz, R. Newton (eds.), London, della Soprintendenza ai beni Artistici e Storici di 115-131. Venezia, 9, 34-58. GORGONI C., LAZZARINI L. and PALLANTE P. (2002b) LAZZARINI L., PENSABENE P. and TURI B. (1999) — — New archaeometric data on «Rosso Antico» Isotopic and petrographic characterization of and other red marbles used in antiquity. In Marmor Lesbium, island of Lesbos, Greece. In «ASMOSIA VI, Interdisciplinary Studies in «Archeomateriaux, marbres et autres roches», Ancient Stone», L. Lazzarini (ed.), Padova, 199- M.Schvoerer (ed.), Bordeaux, 125-129. 206. LAZZARINI L. and CANCELLIERE S. (2000) — GREEN W.A., YOUNG S.M.M., VAN DER MERVE N.J. Characterisation of the white marble of two and HERRMANN J.J. (2002) — Source tracing of unpublished ancient Roman quarries on the marble: trace element analysis with inductively Islands of Fourni and Skyros (Greece). Per. coupled plasma-mass spectrometry. In Mineral., 69, 1, 49-62. «ASMOSIA 5: Interdisciplinary Studies on LAZZARINI L., PONTI G., PREITE MARTINEZ M., Ancient Stone», J.J. Herrmann, N. Herz, R. ROCKWELL P. and TURI B. (2002) — Historical, Newton (eds.), London, 132-142. technical, petrographic and isotopic features of HERZ N. (1955) — Petrofabrics and classical Aphrodisian marble. In «ASMOSIA 5: archaeology, Am. J. Sci., 253, 299-305. Interdisciplinary Studies on Ancient Stone», J.J. HERZ N. (1988) — The oxygen and carbon isotopic Herrmann, N. Herz, R. Newton (eds.), London, database for classical marble. In «Classical 163-168. marble: geochemistry, technology, trade» N. LAZZARINI L. and ANTONELLI F. (2003) — Herz, M. Waelkens (eds.), Dordrecht, 305-314. Petrographic and isotopic characterization of the HERZ N. (1987) — Classical marble quarries of marble of the island of Tinos (Greece), Thasos. In «Antike edel-und Buntmetallgewinnung Archaeometry 45, 541-552. auf Thasos», Bochum, 232-240. LEPSIUS R. (1890) — Griechische marmorstudien, HERZ N. (2000) — The classical marble quarries of Berlin. Paros: Paros I, II and III. In «Paria Lithos: Parian LLOYD R.V., SMITH P.W. and HASKELL H.W. (1985) quarries, marble and workshops of sculpture», — Evaluation of the manganese EPR method of D.U.Schilardi, D.Katsonopoulou (eds.), Athens, marble characterisation. Archaeometry, 27, 108- 27-32. 116. HERZ N. and DEAN N.E. (1986) — Stable isotopes MANFRA L., MASI U. and TURI B. (1975) — Carbon and archaeological geology: the Carrara marble, and oxygen isotope ratios of marbles from some northern Italy. Applied Geochemistry, 1, 139-151. ancient quarries of Western Anatolia and their HERZ N. and PRICHETT W.K. (1953) — Marble in archaeological significance, Archaeometry 17, Attic epigraphy, Am. J. Sci., 57, 71-83. 215-221. KOLLER K., DEPAEPE P. and MOENS L. (in press) — MARINOS G.P. (1948) — Notes on the structure of The Ephesian marble quarries. Topography, Greek Marbles, Am. J. Sci., 246, 386-389. analysis, conclusions. In ASMOSIA VII, MARIOTTINI M. (1998) — La provenienza dei marmi «Interdisciplinary studies on ancient stone», Y. cristallini usati in antico: un problema aperto. In Maniatis (ed.), Athens. «Marmi Antichi II, cave e tecnica di lavorazione, KRETZ R. (1966) — Grain size distribution for provenienze e distribuzione», P. Pensabene, certain metamorphic minerals in relation to (Ed.), Roma, 23-35. nucleation and growth, Geol. 74, 147. MATTHEWS K.J. (1997) — The establishment of a LAZZARINI L. and MARIOTTINI M. (1987) — La data base of neutron activation analyses of white provenienza dei marmi cristallini usati in antico: marble. Archaeometry, 39, 321-332. un nuovo contributo al problema dal rapporto MCCREA J.M. (1950) — On the isotopic chemistry Archaeometric aspects of white and coloured marbles used in antiquity: the state of the art 125

of carbonates and a paleotemperature scale. J. PIKE S. (1999) — Preliminary results of a systematic Chem. Phys., 18, 849-857. study of Mount Pentelikon, Attica, Greece. In MELONI S., ODDONE M. and ZEZZA U. (1995) — «Archeomateriaux, marbres et autres roches», M. Rare-earth element patterns of white marbles Schvoerer (ed.), Bordeaux, 165-170. from ancient quarries in Carrara (Italy). In «The POLIKRETI K. and MANIATIS Y. (2002) — A new study of marble and other stones used in methodology for the provenance of marble based antiquity», Y. Maniatis, N. Herz, Y. Basiakos on EPR spectroscopy. Archaeometry, 44, 1, 1-21. (eds), London, 181-186. RENFREW C. and SPRINGER PEACY J. (1968) — MOENS L., ROOS P., DE RUDDER J., DE PAEPE P., Aegean marble: a petrological study, Annual of VAN HENDE J., MARECHAL R. and WAELKENS M. the British School at Athens 63, 45-66. (1988) — A multi-method approach to the RYBACH L. and NISSEN H.U. (1964) — Neutron identification of white marbles used in antique activation of Mn and Na traces in marbles worked artifacts. In «Classical marble: geochemistry, by the ancient , in Proceedings of technology, trade» N. Herz, M. Waelkens (eds.), Radiochemical Methods of Analysis, Salzburg, Dordrecht, 243-250. 105-117. MONNA D. and PENSABENE P. (1977) — Marmi BIRICOTTI F. and SEVERI M. (2004) — A non- dell’Asia Minore, Roma. destructive methodology for the characterization PAPAGEORGAKIS I.E. (1967) — Die in der of white marble of artistic and archaeological Marmorindustrie nutzbaren Gesteine interest. J. Cult. Her. (in press). Griechenlands (in Greek), Annales Gèologiques SPRY A. (1976) — Metamorphic Texture, Oxford. des Pays Ellèniques, 18, German abstract 225-30. VINOGRADOV A.P., RONOV A.B. and RATYNSKI V.M. PENTIA M., HERZ N. and TURI B. (2002) — (1952) — Changes of chemical composition of Provenance determination of classical marbles: a carbonate rock of the Russian platform, Izv. statistical test based on 87Sr/86Sr, 18°/16° and Akad. Nauk. SSSR Geological Service 1, 33-50. 13C/12C isotopic ratio. In «ASMOSIA VI, YOUNG W.J. and ASHMOLE B. (1968) — The Boston Interdisciplinary Studies in Ancient Stone», relief and the Ludovisi Throne, Bulletin of the L.Lazzarini (ed.), Padova, 219-226. Museum of Fine Arts 66, Boston, 124-166. PESCHLOW-BINDOKAT A. and GERMANN K. (1981) — WEISS L.E. (1954) — Fabric analysis of some Greek Die Steinbrueche von Milet und Herakleia am marbles and its applications to archaeology, Am. Latmos. JdJ, 96, 157-235. J. Sci., 252, 641-662.